1. Field of the Invention
The present invention generally relates to compositions used to prevent ice formation and, more particularly, to a synthetic polypeptide which lowers the freezing point of water and a synthetic gene for producing the polypeptide in bacteria.
2. Description of the Prior Art
Ice formation is an ever present problem in today's world. Ice on roadways and bridges is believed to be the cause of numerous automobile accidents. Ice on aircraft wings is known to decrease lift and increase the weight of the aircraft and is believed to be responsible for causing at least one jumbo jet to crash land. Ice continually forms on the decks of boats which sail in arctic and antarctic waters, thereby making operations aboard the boats extremely hazardous. Ice caused during severe weather can also result in millions of dollars worth of damage to crops such as grapefruits and oranges.
Ice prevention in the roadway, bridge, and boat environments typically involves applying large amounts of toxic chemicals such as glycols and salts to ice formed on the surfaces of the roadway, bridge or boat. Application of glycols and salts is reactive in nature since the conditions for ice formation must all ready be present when the glycols and salts are applied. The chemicals used for ice prevention can be harmful to the environment and tend to cause rust formation on automobiles which travel over the salt covered roadways. Ice prevention in aircraft usually involves onboard inflatable members and heating elements which break and melt the ice after it forms. Such ice protection schemes require energy and are only suitable for specific applications. Presently, there is no generally accepted ice prevention scheme used in agriculture.
Efforts have been made to find less toxic alternatives for ice prevention which are suitable for ice prevention/suppression in a wide variety of applications. U.S. Pat. Nos. 4,045,910 and 4,161,084 to Arny et al. disclose protecting plants from frost damage by applying non-ice nucleating bacteria to the plants before the onset of freezing cold. The non-ice nucleating bacteria are supposed to compete with native ice nucleating bacteria and prevent ice formation by reducing the number of potential "triggers" to crystallization. One drawback of the Arny et al. ice prevention method is that it involves the release of genetically modified bacteria into the environment. U.S. Pat. No. 4,484,409 to Caple et al. discloses chemically synthesizing polymeric ice nucleation inhibitors via free radical polymerization. The polymers produced in Caple et al. have a tightly controlled spacing of about 15 Angstroms (.ANG.) between the hydrophobic and hydrophilic groups. The polymers are sprayed on the plants and are designed to inhibit ice formation. U.S. Pat. No. 4,834,899 to Klevecz discloses applying a bactericide to plants to prevent frost damage by killing the ice nucleating bacteria. U.S. Pat. No. 4,601,842 to Caple et al. discloses applying a proteinaceous material obtained from cold weather plants to growing crops for protection from frost damage.
Recently, naturally produced antifreeze materials that are present in cold water fish have been investigated for their utility as ice supressors (see, for example, Pickett et al., Eur. J. Biochem. 143:35-38 (1984)). These materials are either peptides or glycopeptides and have been designated as AFPs for "antifreeze peptides" and AFGPs for "antifreeze glycopeptides", respectively. The AFPs and AFGPs are produced by the fish to prevent the formation of ice in their body fluids so that they may survive in water temperatures below freezing. There are several subclassifications of AFPs and AFGPs, with the simplest being the alanine rich peptides found in several species of flounder.
FIG. 1 shows the primary amino acid sequence of a native AFP isolated from the winter flounder Pseudopleuronectes americanus (see, DeVries, Phil. Trans. of the Roy. Soc. of London (series B) 304:575-588 (1984)). This is a class 1 or alanine rich AFP. There are 38 amino acids (SEQ ID NO:1) in the AFP with the last 33 amino acids comprising essentially 3 repeats of 11 amino acids (i.e., there are slight differences in the repeating groups where, for example, leucine (leu) is substituted for alanine (ala) and another polar amino acid (xaa) can be substituted for aspartic acid (asp)). The non-polar amino acids, i.e., ala and leu, are used primarily as "spacers" which position the polar amino acids, i.e., thr and asp, approximately 4.5 angstroms (.ANG.) apart. The AFP of SEQ ID NO:1 is but one example of a wide variety of AFPs. Most naturally occurring class 1 AFPs are approximately 70% ala and include three to five repeats of a sequence which includes two amino acids capable of hydrogen binding, i.e., thr and asp, spaced approximately 4.5 .ANG. apart. Chakrabartty et al., in Journal of Biological Chemistry 264:11307-11312 (1989) and 264:11313-11316 (1989), disclosed the direct chemical synthesis of AFPs. It was determined in Chakrabartty et al. that the minimum size for activity of an AFP is 3 repeats of the sequence. There are no known naturally produced AFPs that include greater than 5 repeats of the sequence.
FIG. 2, which is taken from the above cited DeVries article, shows an AFP interacting with an ice crystal during ice formation. As shown in FIG. 2, the secondary structure of a class 1 AFP is an alpha helix and the tertiary structure is a straight rod. The polar amino acids, thr and asp, are positioned on one side of the rod and form hydrogen bonds with the water molecules along the .alpha.-axis of the ice crystal. The 4.5 .ANG. spacing of the polar amino acids in the AFP is precisely the spacing of water molecules in a forming ice crystal.
FIGS. 3a and 3b, which are also taken from the above-cited DeVries article, shows that ice crystals form in two directions. The crystals grow outward along the .alpha.-axis and at right angles to this plane along the c-axis. By binding to the forming prism faces in the .alpha.-axis, the AFPs disrupt the formation of the step, thereby preventing smooth formation of the face, and resulting in curved fronts. Site 10 shows an area where two AFP molecules have bound closely together which has resulted in an even more effective blockage of ice crystal growth. The same extra blockage effect would occur with a longer molecule that can block more of the step.
The mechanism of ice crystal growth suppression by AFPs illustrated in FIGS. 2 and 3b falls under the "adsorption inhibition hypothesis" proposed by DeVries. As illustrated, the ice crystal can only grow in unblocked regions because the presence of the AFP ties up the potential insertion sites for new water to come into the lattice. In theory, since all water adds to the lattice by hydrogen bonding too, if all possible active sites are bound up, the AFPs must by moved out of the way before new water can be added. Attachment of the AFP to the ice crystal increases the ratio of molecular volume to surface area, and in order for freezing to occur, more energy must be removed from the system than would be required in the absence of the AFPs. Hence, the freezing point of water is lowered by the binding action of AFPs. The mechanism of action of AFPs is somewhat analogous to "poisoning" the growth of a crystal by the presence of an impurity where the AFP acts as the impurity. Melting point, however, is not lowered, and this phenomenon, which has been known for many years, is referred to as "thermal hysteresis". Freezing point depression is not a colligative reaction, i.e., very low concentrations of AFP in pure solutions are known to have approximately five hundred times greater depression than colligative processes would predict.
Ideally, AFPs could be used to suppress ice formation in a wide variety of environments. AFPs have the advantage that they can be applied to a road surface or to an agricultural plant ahead of time so that they would interact with ice during formation and, further, they can be applied after the onset of ice formation and serve to prevent continued ice crystal formation. In addition, since AFPs are simply polypeptide chains, they pose no hazard to the environment. In winter flounder, the concentration of AFPs range from 1.0% to 3.0% depending on the species and the season; hence, AFPs are not produced in large enough quantities in fish for the fish to be harvested as a source for an ice preventing agent. Moreover, AFPs and AFGPs are typically only produced in the fish during the winter months. As noted above, Chakrabartty et al. have shown it is possible to synthesize AFP using direct chemical processes; however, these processes can be expensive and time consuming. Peters et al., in Protein Engineering 3:145-151 (1989), disclosed producing a semisynthetic winter flounder AFP in Escherichia coli (E. coli). In Peters et al, a gene constructed of a fused synthetic deoxyribonucleic acid (DNA) fragment and a DNA fragment derived from a full length winter flounder clone was inserted into a plasmid and the plasmid was placed in the E. coli for production of a fusion protein. The biosynthetic fusion protein produced contained part of a pro-AFP and part of a .beta.-galactosidase peptide and had limited antifreeze activity after cleavage from .beta.-galactosidase. What is needed is a synthetic polypeptide which has greater freezing point depression capability than those AFPs which occur in nature and which can be economically produced by biosynthetic processes.